Time series analysis of government information about COVID-19 on social media and the number of new cases during the first 6 months of the pandemic: the Brazil case

• Autor(a) principal: Larissa Tavares da Silva
• Data de Publicação: 2021
• Outros Autores: Angélica Maria Cupertino Lopes Marinho, Nayara Santos Braga, Thiago Rezende Dos Santos, Mauro Henrique Nogueira Guimarães de Abreu, Cristiane Meira Assunção, Fernanda Moraes Ferreira
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UFMG
• Texto Completo: https://doi.org/10.33448/rsd-v10i16.23797; http://hdl.handle.net/1843/58525; https://orcid.org/0000-0002-0961-0900; https://orcid.org/0000-0003-3107-0223; https://orcid.org/0000-0001-7391-0409; https://orcid.org/0000-0003-2244-2718; https://orcid.org/0000-0001-8794-5725; https://orcid.org/0000-0002-1464-7812; https://orcid.org/0000-0001-9400-1167
• topic: COVID-19; Health communication; Health education; Access to information; Social networking; COVID-19; Comunicação em saúde; Educação em saúde; Acesso à informação; Rede social

This retrospective study sought to assess the association between the evolution of the COVID-19 pandemic in Brazil and the quality of educational materials published in the official profiles of Brazilian government health agencies on Instagram. Posts about COVID-19, published between January 31 and August 15, 2021, were selected, dated, quantified and classified according to their content by three researchers. Public’s engagement was calculated by the number of likes, comments and views. The quality of the educational posts was assessed by two trained and calibrated researchers (Kappa intra and inter-examiners, k=0.96 and k=0.92, respectively), using the Brazilian version of the Clear Communication Index (BR-CDC-CCI), the number of new COVID-19 cases was collected using the COVID-19 epidemic calculator provided by PAHO at https://covid-calc.org/. The relationship between the evolution of the COVID-19 indicator and the quality of educational posts was calculated using the statistical model of a fortnightly time series. On average, educational posts reached 6.4 in the BR-CDC-CCI score (median = 6.5). In the multiple model adjusted for the amount of educational posts and public engagement, it was observed that for each increase of one point in the BR-CDC-CCI score, there was a reduction of 327,864 new cases of Covid-19 (p <0.001). It was concluded that there was a relationship between the low quality of posts and the greater number of new cases of the disease, indicating the need for greater attention from Brazilian government agencies with the quality of information made available on social networks to help control the COVID-19 pandemic.